Modelling the fatigue life of polymer-matrix fibre-composite components

Citation
O. Attia et al., Modelling the fatigue life of polymer-matrix fibre-composite components, COMP SCI T, 61(15), 2001, pp. 2273-2283
Citations number
19
Categorie Soggetti
Material Science & Engineering
Journal title
COMPOSITES SCIENCE AND TECHNOLOGY
ISSN journal
02663538 → ACNP
Volume
61
Issue
15
Year of publication
2001
Pages
2273 - 2283
Database
ISI
SICI code
0266-3538(2001)61:15<2273:MTFLOP>2.0.ZU;2-E
Abstract
A methodology has been proposed for predicting the fatigue life of fibre-co mposite components and structures which combines relatively short-term frac ture mechanics data, obtained from experimental measurements, with a finite -element analysis (FEA) of the component or structure. The approach has bee n used to study the growth of damage in, and the cyclic fatigue life of I-b eams. The beams were manufactured using carbon-fibre-reinforced-plastic (CF RP) and contained a 60 mm diameter notch in the web. Experimental work has shown that the development of significant damage was limited to the region of material in the web around the 60 mm diameter notch. A significant amoun t of matrix micro-cracking damage occurred within the first 0.5 x 10(6) fat igue cycles, mainly in the + 45 degrees plies and 0 degrees plies, in which the fibres are orientated at 90 degrees and at 45 degrees, respectively, t o the local tensile stress. This matrix cracking eventually led to some lim ited delamination after about 0.5 x 10(6) cycles had elapsed and this occur red mainly along the global +45 degrees/-45 degrees ply interfaces, with so me delamination also occurring along the +45 degrees /0 degrees ply interfa ce. The growth of these two types of damage eventually led to fibre fractur e which was the final cause of structural failure of the web material, and hence of the I-beam. Thus, the model has concentrated upon modelling these types of damage mechanisms. The agreement between the results from the theo retical model and the experiments is good, especially when it is considered that there are no 'adjustable factors' involved in the modelling studies. For example, the theoretical model predicts the number of cycles, Nf, for t he I-beam to fail structurally to be about 4.9 x 10(6) cycles. The experime ntally measured value was 4.78 x 10(6) cycles. (C) 2001 Elsevier Science Lt d. All rights reserved.